Previous motor actions outweigh sensory information in sensorimotor statistical learning

Abstract

Humans can use their previous experience in form of statistical priors to improve decisions. It is however unclear how such priors are learned and represented. Importantly, it has remained elusive whether prior learning is independent of the sensorimotor system involved in the learning process or not, as both modality-specific and modality-general learning have been reported in the past. Here, we used a saccadic eye movement task to probe the learning and representation of a spatial prior across a few trials. In this task, learning occurs in an unsupervised manner and through encountering trial-by-trial visual hints drawn from a distribution centered on the target location. Using a model-comparison approach, we found that participants’ prior knowledge is largely represented in the form of their previous motor actions, with minimal influence from the previously seen visual hints. By using two different motor contexts for response (looking either at the estimated target location, or exactly opposite to it), we could further compare whether prior experience obtained in one motor context can be transferred to the other. Although learning curves were highly similar, and participants seemed to use the same strategy for both response types, they could not fully transfer their knowledge between contexts, as performance and confidence ratings dropped after a switch of the required response. Together, our results suggest that humans preferably use the internal representations of their previous motor actions, rather than past incoming sensory information, to form statistical sensorimotor priors on the timescale of a few trials.

Significance Statement

Humans can learn statistical regularities and later use them as priors to inform decisions. It remains unclear what type of representation is used to store and integrate past experience. We designed an experiment where humans had to combine visual information over multiple trials to locate a hidden target location. Using computational modelling, we found that participants represented past experience in the form of their previous decisions, and not directly by memorizing the visual cues. As a consequence of overweighing past decisions relative to the veridical visual information, gained experience did not generalize across two different contexts, albeit they differed minimally with respect to the prior. Hence, the process through which past experience is learned determines its influence on our decisions.